Table 8 Performance comparison of the system (Unit: %).
Evaluation Metric | Proposed System | Grammarly | GPT-3.5 Fine-tuned Model |
|---|---|---|---|
Verb Tense F1 | 93.4 | 79.2 | 85.6 |
Article Misuse Recall | 91.7 | 83.5 | 81.9 |
Subject-Verb Agreement Precision | 96.3 | 88.7 | 92.1 |
Average Feedback Latency (sec) | 1.8 | 2.9 | 3.5 |
User Satisfaction (5-point scale) | 4.5 | 4.1 | 3.9 |
